The Temple of Memories: History, Power, and Morality in a Chinese Village 9780804764926

This study focuses on the politics of memory in the village of Dachuan in northwest China, in which 85 percent of the vi

172 117 92MB

English Pages 232 [240] Year 2022

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Temple of Memories: History, Power, and Morality in a Chinese Village
 9780804764926

Citation preview

The Temple of Memories

The Temple of Memories HISTORY, POWER, AND MORALITY IN A CHINESE VILLAGE

JUN JING

Stanford University Press Stanford, California

Stanford University Press Stanford, California © 1996 by the Board of Trustees of the

Leland Stanford Junior University Printed in the United States of America CIP data appear at the end of the book

For my parents and Jeanne

Acknowledgments

My debt to James L. Watson is immeasurable. As my graduate advisor at Harvard University, Professor Watson first encouraged me to work in northwest China to investigate the problem of social memory. Professor Watson is a wonderful mentor and I owe him the deepest gratitude. I am also deeply grateful to Arthur Kleinman and K. C. Chang at Harvard. Professor Kleinman has been a constant intellectual stimulus. Professor Chang's unflinching sense of humor and optimism have raised my spirits on more than one occasion. My intellectual journey at Harvard could not have been so enjoyable and rewarding an experience without the encouragement of other teachers, especially Thomas Barfield, Kenneth George, Michael Herzfeld, Roderick MacFarquhar, David Maybury-Lewis, Sally Moore, Pauline Peters, Parker Shipton, Stanley Tambiah, Rubie Watson, Nur Yalman, and Robin Yates. As conscientious educators and accomplished scholars, they offered a sympathetic ear and a guiding hand. They occupy a special place in my heart. Many people offered invaluable suggestions after reading portions of the manuscript or listening to ideas in gestation. There is no way I can render unto them as much as I have received. A few who deserve particular mention are Helen Siu at Yale Univii

viii

Acknowledgments

versity, Robert Weller at Boston University, and Yan Yunxiang at the University of California, Los Angeles. And when the business of writing became too demanding, several friends extended logistical and emotional support. They include Sangmee Bak, Maris Gillette, Lida Junghans, Matthew Kohrman, Lu Qiwen, and Wang Daw-hwang. Numerous funding and academic institutions provided "the iron rice bowl" that made this study possible. The Institute of Sociology at Beijing University gave me my first opportunity to work in northwest China during the summer of 1989. A Mellon Fellowship for Summer Research enabled me to travel from the United States to China's Gansu province in 1991. A semester of predissertation research in 1991 was supported by a Peabody Merit Scholarship. A 1992 Foreign Languages and Area Studies Fellowship supported my library research on Gansu and training in the local dialect. The National Science Foundation and the Wenner-Gren Foundation for Anthropological Research financed my fieldwork in 1992. The Beijing Social and Economic Survey Center served as the institutional sponsor for my fieldwork. My archival research in Taiwan in 1993 was aided by another Mellon Fellowship. The dissertation write-up was funded by a Charlotte W. Newcombe Doctoral Dissertation Fellowship of the Woodrow Wilson National Fellowship Foundation. I am grateful to all these organizations for their support. At a critical point of revision, Myron Cohen at Columbia U niversity carefully read the manuscript. His many insightful suggestions for improvement are immensely appreciated. I also owe a great debt of gratitude to Muriel Bell, my editor at Stanford University Press. Her enthusiastic support and editorial guidance carried me through the final stage of writing. I convey special thanks to my parents, Jing Ruiyin and Li Yunpu, for having faith in my work. My deepest appreciation goes to my wife, Jeanne Moore, who has extended the boundaries of my compassion.

Contents

I.

Introduction: A Study of Social Memory

I

2. Memory of Historical Possibilities

23

3· Memory of Revolutionary Terror 4· Memory of Communal Trauma

45 69

s.

Memory of Local Animosity

87

6. Memory of Ritual Language

IOI

7· Memory of Genealogical Retainers 8. Memory of Cultural Symbols

IIS

9· Finding Memories in Gansu Notes

I77

References

I87

Character List 205 Index 207 Photographs follow p. 86

144 163

Maps

r. Location of Gansu province in the People's Republic of China

4

Lanzhou, Yongjing county, and Dachuan village

6

2.

The Temple of Memories

1

Introduction: A Study of Social Memory

A Village Destroyed The Kongs of Dachuan cannot forget that winter, more than three decades ago, when their village was effaced and life as they had known it ended. For much of 1960 they had ignored, then resisted, the government's declaration that their homes lay in the path of one of the more ambitious projects of the Great Leap Forward, and that by autumn's end they would have to make way for a hydroelectric dam and reservoir.1 But how could they, as Kongs who saw themselves as descendants of the ancient sage Confucius (kong fu zi in Chinese), abandon their ancestral tombs and the land that had been their haven for centuries? And, in any case, how, as one of them put it, could the mighty Yellow River really be CUt off at its waist"-a goal that had defeated the past efforts of Chinese and American engineers ?2 So months lapsed, the deadline passed, and still the Kongs stayed on. And then, on a chill December night, the militia entered, shock troops of eviction, targeting first households without strong young men. Old women screamed and clung to their beds, refusing to leave. They were carried out bodily. The supporting pillars of their houses were roped to mules and pulled 11

I

2

The Temple of Memories

down. As dawn broke, the frightened villagers began dismantling their own houses in a scramble to salvage what they could to build shelter elsewhere. They hastily dug up the graves of immediate ancestors and close relatives, and, in violation of all tradition, unceremoniously threw bones in cement sacks or whatever other containers they could find for reburial on higher ground. "It was no time for being proper about such things," an elderly villager recalled years later. Nor did they have the physical strength to save older graves; the trauma of dislocation was exacerbated by a debilitating famine, the worst in modern Chinese history.3 Under the double burden of hunger and shock, half the local residents began resettling into other nearby communities while the remaining population retreated a short distance uphill. In March 1961, the Yellow River, China's second largest, was stopped just downstream from Dachuan, its progress halted by a dam 57 meters high. The river was compelled, by design, to surge sideways to create a reservoir behind the dam. By the time the dam's floodgate was lifted 48 hours later, much of Dachuan was submerged, forever buried under the reservoir's waters. The destruction of their village was the central event in a long procession of tragedies for the Kongs under the new Communist regime. Before r 949, the Kongs had dominated this part of northwest China's Gansu province. By status and name, they personified the old order that the Communist government had vowed to supplant. Throughout the Maoist era (1949-76), the Kongs were hit hard by assaults on the vestiges of old China, culminating, in their case, with the 1973-74 Campaign Against Lin Biao and Confucius.4 Only in the mid-198o's, with the gradual relaxation of political controls, did some of the Kongs consider it possible to reassert themselves and work toward some form of redress and even healing. My own path converged with that of the Kongs in the summer of 1989, weeks after the military suppression of demonstrations in Beijing's Tiananmen Square. At the time, I was a researcher at Beijing University's Institute of Sociology, directed by there-

A STUDY OF SOCIAL MEMORY

3

nowned anthropologist Fei Xiaotong. Fei had earlier suggested as a research topic looking into the long-term consequences of population dislocation caused by water projects in northwest China. In fact, the institute's initial findings had already figured in the debate over building a dam at the Yangtze River's Three Gorges. With the university effectively shut down, first by the student protests and then by the political crackdown, and with little interest in remaining in Beijing, where friends and colleagues were being arrested, several of us decided to resume that research. Thus, in July 1989 I set off with another researcher and two graduate students for Gansu and my first encounter with the Kongs in the relocated village of Dachuan. At first, I expected this to be a short-term study, mostly about reservoir resettlement. Instead, I returned to Dachuan, after entering the doctoral program in anthropology at Harvard University, to spend the summer of 1991, eight months in 1992, and brief stays in 1993 and r 99 s, learning not only about resettlement, but about other aspects of the local experience of radical socialism during the Maoist era and about the terms of life in the post-Mao period. The 1989 trip was to prove to be my first field research into the range of issues generally addressed by scholars under the heading "social memory." The ultimate goal of social memory research is to investigate the transformative impact of grouplife requirements and collective interests upon the overall framework and specific contents of personal recollections. As explained below, Dachuan is an ideal place for this kind of research.

The Regional Setting of Fieldwork The village of Dachuan is about 8o kilometers southwest of Lanzhou, Gansu's provincial capital. On the map of China, Gansu is a long, diagonal wedge, bounded by Shaanxi on the east, Sichuan on the south, Xinjiang on the west, the Qinghai-Tibet highlands on the southwest, and Ningxia and Inner Mongolia on the north. In other words, Gansu forms the narrow corridor

The Temple of Memories

4

'

'

''

XINJIANG

. ._ J--

....

-- .... } ' ' '

I

(

QINGHAI '-,

TIBET

,_....

"

_.J...

... ,__

\\

'

r. Location of Gansu province in the People's Republic of China.

linking China's heartland to Inner Asia, a vast terrain settled by Tibetans, Mongolians, Uighurs, Chinese-speaking Muslims, and many other ethnic groups. For centuries, China's attempts to control its northwest frontiers have relied on the passageway provided by Gansu. Given its strategic importance, it is not surprising that the province's capital originated as a military outpost in 86 B.C. When this outpost became, in A.D. 58 I, the seat of a prefectural government (Wei 1988: 9), it was named Lanzhou, or "Mount Lan Prefecture." Mount Lan is actually a range of loess hills south of the once-walled city. Hemmed in by these hills on one side and the Yellow River on the other, the city has by now grown into an hourglass-shaped metropolis. Its commercial center in the north and its industrial district in the south are linked by a road about ten kilometers long. Lanzhou was a major center of caravan traffic until World War

A STUDY OF SOCIAL MEMORY

5

II. It has since been linked by rail to Xinjiang, Qinghai, and Inner Mongolia. It is also connected by rail to an important depot in Qinghai, which serves as the staging point for trucks and buses headed for Lhasa. In r 948, the city had r 6o,ooo residents (China Handbook Editorial Board 1950: 43). The population now exceeds two million. Rapid industrialization since the Communist takeover in r 949 has turned Lanzhou into northwest China's second-largest industrial center, after Xian. From the start, its industrial programs emphasized the province's mineral resources and, consequently, heavy industry. This emphasis also set the pattern for the industrialization of the province's smaller cities. Rich deposits of coal and iron ore provided the base of Gansu's ferrous metal industry, which produces pig iron, steel, steel alloys, and ferrosilicon. Abundant copper, nickel, lead, and zinc deposits supplied its nonferrous metal industry (World Bank r 9 8 8: 40). These heavy industry projects have benefited from the construction of large-scale dams and electric-generating plants on the upper reaches of the Yellow River. Over 6 s percent of Gansu's 20 million people lived in rural areas in 1992. Its rural population had, in 1984, the lowest per capita income in China-221 yuan, compared with a national average of 356 yuan (World Bank 1988: r). In the early 1990's, its rural population remained the nation's poorest (State Statistics Bureau 1992, 1993a). This persistence of poverty owes much to Gansu's natural environment: a short growing season, sparse rainfall, and severe soil erosion. Social conditions also have contributed to poverty. In the first half of the twentieth century, Gansu was plagued by official corruption, relentless military conscription, ethnic tensions, and almost constant warfare (see Ekvall 1938: 64-69; Lipman 1980, 1984:285-316, 1990: 65-86; Perry 1983: 355-82). Under the collective farming system, from 1958 to 1981, deforestation, population resettlement, aggressive efforts to expand farmland, and major hydroelectric projects combined to aggravate Gansu's ecological depredation (Greer 1979; Jing 1989; Smil 1984: 10-61;

The Temple of Memories

6

~ hydroelectric dam

2.

Lanzhou, Yongjing county, and Dachuan village.

Wang I989). Since the mid-I98o's, this province has been a major recipient of special poverty-relief loans from the central government.

Basic Information About Dachuan Village Dachuan, which means "big valley," is under the jurisdiction of a rural county named Yongjing. The village is perched at the midpoint of an S-shaped valley carved by the Yellow River as it cascades down from the Qinghai-Tibet plateau. Its residential area is clustered on the slope between the reservoir built in I 96 I and a barren mountain range. A I992 village census recorded 3,3Io residents and 638locally registered households. Most of Dachuan's residents depend on farming for their basic livelihood; their chief crops are wheat, maize, and potatoes. However, reservoir construction and population growth have reduced the

A STUDY OF SOCIAL MEMORY

7

village's per capita farmland to o.6 mu (one mu is about onesixth of an acre). This extreme!y limited amount of farmland has made it virtually impossible for villagers to meet their basic needs solely through agriculture. Dachuan is nonetheless better off than many nearby communities, thanks to the villagers' persistent search for seasonal jobs in cities and at state-run construction sites. And from the early 198o's to 1992, local people dug more than roo fish ponds from the salty marshes on the edge of the reservoir. These fish ponds, along with the outside jobs, are important sources of cash income. Even so, the village is mired in extreme poverty. When I revisited Dachuan in September 1993, its chief administrator calculated that nearly 40 percent of the local households were still living below the official poverty line, set in Yongjing at an annual per capita income of 300-400 yuan, or $so-$6s by the official exchange rates that year. In its outward appearance, Dachuan resembles many other villages in this part of Gansu. The typical dwelling is set within a walled courtyard. The central hall invariably faces south, flanked by perpendicular side rooms facing east and west. The buildings are mostly made of rammed earth walls and wooden pillars, which support roofs that are generally tilted slightly downward toward the courtyard. In contrast to the barren and windswept mountains behind the village and across the river, the settlement is encircled by thickets of poplars planted along dirt paths and between fields. Apple and pear trees are grown within walled courtyards and enclosed gardens. Reforestation has been a major achievement in Dachuan since the termination of the collective farming system in 1981. The village boasted 200 mature trees per household in 1992, all privately owned.

Local Kinship Structure Interpersonal relations in Dachuan are tightly knit by descent and marriage. Although it is not a single-surname village, 8 s percent of the local households in 1992 were surnamed Kong. The

8

The Temple of Memories

balance comprised r6 other surnames. Except for a group surnamed Li, whose ancestors had settled in Dachuan earlier than the Kongs, the others came to the village as refugees from war and famine in the late nineteenth century and during the first quarter of the twentieth century. In 1992, over 25 percent of the married women in the other surname groups were Kongs. Dachuan also has been the center of a large ancestral cult, whose members, roughly 2o,ooo people in 1992, are surnamed Kong and are distributed over 23 villages within Yongjing county. These Kongs trace their ancestry to Confucius through a Guangdong-born ancestor who migrated to Gansu and settled in Lanzhou about six centuries ago. He is said to have been related to the Guangdong branch (ling nan pai) of the nationwide Kong clan headquartered in Confucius's hometown of Qufu, in the coastal province of Shan dong. The genealogical claim of the Kongs in Yongjing was not formally recognized by the Qufubased council of clan elders until 1937, when the Confucius Clan Genealogy (kong zi shi jia pu) was updated and expanded to include not only the Kongs in Qufu but also those scattered throughout China.5 Despite the rather late recognition by Qufu, the Kongs in Dachuan and adjacent Kong villages had long identified themselves with the Confucian heritage by three means: a carefully guarded collection of genealogical records, a cycle of ancestral rituals, and a temple dedicated to Confucius. It was around the year r soo that a group of Kongs originally living in a village near Lanzhou moved to Dachuan. Since the four men heading this group were brothers and were buried in the village after death, the Kongs of Yongjing consider Dachuan their "old settlement" (lao zhuang). From a historical point of view, the social organization of the Kongs in Yongjing constituted what historians and anthropologists specializing in China call "a higher-order lineage" (see H. Baker 1979: 67-70; Freedman 1966: 19-21, J. Watson 1982b: 6o8-9). The nucleus of this lineage was Dachuan's Confucius temple-a corporate estate and a fixed site at which its members could celebrate ritual unity. The institutional framework of this multi village lineage

A STUDY OF SOCIAL MEMORY

9

broke down under the blows of Maoist campaigns, but, as we shall see, its recovery was evident by the mid-198o's. Because they trace local history to the four brothers who settled and then died in Dachuan, the Kongs in the valley of Yongjing county have been divided into four sublineages locally known as the "four gates" (si men). Population growth, economic pressures, and reservoir construction have forced many Kongs to leave Dachuan for other places. However, members of all four sublineages are still found in the relocated village of Dachuan, whereas the Kongs in the other 22 Yongjingvillages tend to be from just one or two of the four sublineages that originated in Dachuan.

Village Administration The Kongs living in Dachuan itself have been further divided into eight smaller divisions. This eight-part structure of agnatic kinship evidently influenced the local implementation of the collective farming system known as "people's communes." In force throughout rural China from 19 58 to the early 198o's, this system had three institutional components. At the lowest level was the "production team." Its leaders were responsible not only for daily job assignments to team members, but for distributing "work points" (gong fen), which were in effect cash or material rewards calculated on the basis of work performance, taking into account team members' sex, age, health, skills, and education. At the middle level was the "production brigade," whose leaders drew up the general agricultural production plan to be implemented by the production teams. Usually, a production brigade comprised several production teams within a village, making brigade leaders a village's most influential cadres. At the top level was the" commune," the administrators of which were appointed by county-level authorities. The administrative body and the Communist Party committee of a given "commune" were in charge of the selection of village-level cadres. The way a "production brigade" was formed in Dachuan was

IO

The Temple of Memories

apparently influenced by what anthropologists call"the social distance of kinship" (see Sahlins I972: I89-204), a principle of political loyalty and economic reciprocity based on differentiated ties between kin. As mentioned above, the Kongs divided themselves into four sublineages, which in Dachuan became eight smaller divisions. In the past, the eight divisions were reflected in clearly demarcated residential areas within Dachuan. They also had separate territories for orchards, farm fields, threshing grounds, and graveyards. When a "production brigade" was organized in Dachuan, the eight Kong divisions within the village formed themselves into eight "production teams" and retained their separate territories. After the village was relocated in I96I to higher ground in the vicinity, farming and residential territories were re-established, primarily along kinship boundaries. In I 98 I, the commune system was ended at Dachuan in favor of agricultural work by individual households. The name "production teams" was changed to "economic cooperatives," which in Dachuan have maintained identifiable boundaries of residential and farming areas, boundaries drawn on the basis of the eightfold divisions of the Kong lineage. Each "economic cooperative" therefore represents a social microcosm within the village's overall structures of residence, administration, property relations, public security, personal ties, and marriages between Kongs and others. Within the agnatic kinship of the Kongs, close relatives trace ancestry through five generations to a common ancestor instead of going all the way back to the four brothers who first settled in Dachuan. This type of agnatic relation is known as the "five mourning grades" (wu fu), a popular reference in Chinese culture to the five kinds of attires required of agnates at funerals (see Wolf I97o: I89-207). In the eyes of Dachuan's residents, all Kongs living in this village and elsewhere in Yongjing are relatives, but only those within the five mourning grades are primary agnates. In theory, these agnates are under the greatest obligation to participate in one another's life-cycle rituals, espe-

A STUDY OF SOCIAL MEMORY

II

cially funerals; they are also expected to provide mutual assistance in times of crisis. The Kongs have long dominated Dachuan both in numbers and in public positions. From I9 so to 1992, the village chief, the local Party boss, the accountant-general, the production team leaders, and the local primary school's principal were all surnamed Kong. It was only in I 99 3 that a Mr. Luo was appointed by higher authorities to take charge of the village's primary school. As in other parts of rural China, Dachuan's local administrators are called "village cadres" (cun gan bu). The administrative power of Dachuan's cadres has declined markedly since the village was decollectivized. But still, they wield a great degree of control when it comes to collecting agricultural taxes, organizing labor for public works, redrawing land allocations to cope with population changes, and implementing the government's birth control policy. In terms of personal wealth and opportunities to take advantage of the entrepreneurial thrust of post-Mao economic reforms, the cadres and former cadres of Dachuan have an unbeatable edge in business situations that require networks of connections, bank loans, travel experience, market information, legal knowledge, and inside word on state policies.6

A Social History of Remembering Although my fieldwork touched on many historical and contemporary aspects of village life in Dachuan, the ethnographic data analyzed throughout this book focus on the rebuilding of a local temple for the worship of Confucius and local Kong ancestors. It will be explained in Chapter 2 why I call this site a "Confucius temple" instead of an "ancestral hall." Rebuilt in 1991, this temple played a pivotal role in the reconstruction of Dachuan's religious life, kinship ideology, and power structure. All specific isues related to this temple will be examined from the theoretical perspective of social memory. These include ritual knowledge, the writing of genealogies, popular images of village history, narratives of suffering, the politics of revenge, conflict-

12

The Temple of Memories

ing perceptions of rituals' significance, and attitudes towards China's experiments with radical socialism. By examining how memories are articulated and transmitted through activities centered around Dachuan's Confucius temple, this study offers a bottom-up approach to the problem of remembrance in a country where mass amnesia and selective remembrance have been vigorously promoted by state authorities. At first glance, the story of Dachuan's Confucius temple may suggest that many traditional ideas and practices prohibited under the first decades of Communist rule have been revived. But a closer look reveals a far more complicated picture. These ideas and practices are not mechanically retrieved from the past; they are blended with cultural inventions, shaped by the local experience of Maoism, and permeated with contemporary concerns. As a landmark of local history, the rebuilt Confucius temple certainly evokes a wide range of memories of this village's preCommunist past, when the Kongs enjoyed a prestige that owed much to their identification with Confucius and their previous political influence and economic status. It is important to bear in mind, however, that the rebuilding of this temple had everything to do with the Kongs' efforts to make sense of their troubled experience under Mao Zedong's rule and the social changes of the post-Mao era. It is thus best to regard the story of Dachuan's Confucius temple, to borrow a phrase from Peter Burke, as "a social history of remembering" (1989: roo). The Dachuan material furnishes an especially interesting basis for formulating cross-cultural comparisons. In a study of how men and women remembered their work lives in an American automobile plant from World War II to 1961, John Bodnar (r989: r2or-2r) noticed several consistent themes in the autobiographical narratives of these retired workers. Commenting on such themes, he pointed out: "Memory was a cognitive device by which historical actors sought to interpret the reality they had lived-and, it appears, they could never do so alone, without reference to a social context" (r989: 1202). One might

A STUDY OF SOCIAL MEMORY

I3

ask: What is this "social context" that narrators of history must invoke? This seemingly simple question has been in fact at the very heart of social memory research from its inception.

The Social Basis of Memory Psychologists began studying memory a century ago, but, with some exceptions, their research was conducted in laboratory settings, using formal experimental techniques to answer theoretical questions about the general principles that govern the workings of memory. In an attempt to endow psychology with scientific respectability, the experimental methods invented by Hermann Ebbinghaus ([ r 88 s] r 964) to test the results of verbal learning were embraced and developed. In typical experiments on memory, stimuli such as nonsense syllables were used as the objects for recall. The human subjects were carefully selected and instructed. The experimental environment was standardized and the amount of time the subjects needed to recall stimuli was painstakingly recorded. These experiments tended to ignore the complex workings of memory in real-life contexts, and they generated few insights into the social basis of remembrance. At a 1976 conference attended mainly by English-speaking psychologists to discuss the practical aspects of memory, Ulric Neisser characterized the inadequacies of research on the social origin of memory by a simple formula: "If X is an interesting or socially significant aspect of memory, then psychologists have hardly ever studied X" (r 97 8: 4). This was a fairly accurate criticism of the state of memory research in psychology up to the late 197o's. The same criticism could have been applied to sociology and anthropology were it not for the valuable work by the French scholar Maurice Halbwachs (r877-1945 ), a disciple and younger colleague of Emile Durkheim. It was in 1925 that Halbwachs developed the concept of "collective memory" and applied it to his studies of how the past is remembered within the milieux of

14

The Temple of Memories

families, religious groups, and social classes. Rejecting various psychological interpretations of memory then gaining currency in Europe, he argued that any discussion of the origin of personal recollections must take into account the impact of such social institutions as kinship, community, religion, political organization, social class, and nation. As an illustration of his central argument, he contended that in stable community life every family has its privately cherished memories and secrets, revealed to none but its own members. This "family memory" (memoire familiale) is not merely a collage of individual recollections. Rather, it is a collective reworking of the past (Halbwachs [1925] 1952:287). There is little doubt that Halbwachs was faithful to the fundamentals of Durkheimian sociology, but he was not a blind follower. That he was an imaginative thinker becomes apparent when one looks at what he added to Durkheim's insights. For example, after developing the concept of "mechanical" and "organic" solidarity in The Division of Labor in Society ([I930] I964), Durkheim focused on the idea of "collective consciousness"-that is, "the body of beliefs and sentiments common to the average of the members of a society" (Aron 1970: IS). For Durkheim, collective consciousness embraces the greater part of individual existence rather than the desire of any particular persons (Aron I 970: I r -24). As he saw it, the prospect of purely individual thought is almost an absurdity, since language and conceptualization develop through social interaction. Phrased differently, the source of collective identity, morality, and religion, in Durkheim's view, is the individual's experience of society, a force greater than oneself and requiring one's allegiance (Douglas I98o: 8). Proceeding from this argument, Durkheim was able to develop the concept of "collective effervescence" and to interpret it as the foundation of human cultural creativity in the spheres of tribal ceremonies, ritual dances, festival meals, or public holidays. In many pages devoted to this concept, he attempted to

A STUDY OF SOCIAL MEMORY

15

demonstrate that cultural creativity is largely rooted in collective enthusiasm, shared sentiments, and unified interests, quite contrary to the then fashionable thought that cultural creativity is the privilege of a handful of individuals. Yet there is a problem with Durkheim's thinking. Even if we assume that societies or groups exhibit cultural creativity and renewal through the celebration of ritual unity in periods of effervescence, one must ask what binds people together in periods of calm, when routine behavior is the order of the day. This problem was successfully addressed by Halbwachs, who argued that it is collective memory that keeps the recall of past events alive in everyday life, making the ritualized re-enactments of history and the excited celebrations of group identity possible over time (see Coser 1992 for a fuller discussion). In retrospect, Halbwachs developed quite original ideas on the issue of memory, best illustrated in three of his works. The first isLes Cadres sociaux de la memoire ([1925] 1952), which lays out his basic theory of collective memory. The second is La Topographie legendaire des evangiles en terre sainte: Etude de memoire collective (I 941 ), a historical study of how Christians, including the Crusaders, employed memories of their religious upbringing to discover, sometimes with great imagination, holy sites during their visits to Jerusalem. The third is La Memoire collective ([1950] 1980), an application of his theory to the analysis of childhood memories, perceptions of time and space, and differences between history and memory?

Three Approaches in Social Memory Research The trailblazing research by Halbwachs did not attract much attention outside France until the 1970's, when it was taken up by historians and social scientists attempting to move beyond the purely psychological discussion of memory. Among recent studies of social memory, at least three basic approaches can be discerned.

16

The Temple of Memories

THE COLLECTIVE MEMORY APPROACH

Following Halbwachs, social historians, anthropologists, and sociologists have tried to investigate what members of a social group share of their past and how (see, e.g., Rosaldo I980; Schuman & Scott I989: 359-8I; Schwartz I99I: 22I-36; Valensi I986: 283-305 ). To this end, Paul Connerton (I989) has explored the importance of ritualized re-enactments of myth and history, while Fran